JACC Vol. 25. No. 5 989 April 1995:989 94

Quinidine Pharmacodynamics in Patients with Arrhythmia: Effects of Left Ventricular Function

ANNE M. GILLIS, MD, L. BRENT MITCHELL, MD, D. GEORGE WYSE, MD, PIJD, FACC, MARGOT McDONALD, BN, HENRY J. DUFF, MD Calgar), Alberta, Canada

Objectives. This study was undertaken to determine whether The QRS concentration-effect relations were not different in the quinidine pharmacodynamics are altered in the presence of left two groups. A significant linear correlation was observed between ventricular dysfunction. the slopes of the concentration-effect relations describing changes Background. Left ventricular function is an independent pre- in QTc intervals and left ventricular ejection fraction (r = 0.7, p < dictor of antiarrhythmic drug efficacy. However, the effects of left 0.001). Nineteen patients with inducible ventricular tachycardia ventricular dysfunction on the pharmacodynamics of antiarrhyth- underwent serial electrophysiologic studies for evaluation of quin- mic drugs have not been studied extensively. idine efficacy. Ventricular tachycardia could not be induced Methods. Signal-averaged electrocardiograms were obtained during quinidine therapy in eight patients. The slopes of the and quinidine plasma concentrations measured during 24-h quin- quinidine concentration-effect relations for QTc intervals were idine washout in 22 patients. significantly higher in quinidine responders than in nonre- Results. Linear quinidine concentration-effect relations were sponders (p < 0.05). observed for QRS and QT intervals corrected for heart rate. The Conclusions. The effects of quinidine on ventricular repolariza- slopes of the concentration-effect relation describing changes in tion are linearly related to left ventricular ejection fraction. the corrected QT (QTc) interval were significantly higher in the Quinidine concentration-effect relations describing ventricular group with left ventricular ejection fraction >0.35 ([mean _+ SD] repolarization are associated with antiarrhythmic efficacy in 29.5 -+ 11.2 ms//~g per ml) than in the group with a low left patients with ventricular tachycardia. ventricular ejection fraction (15.7 -+ 9.7 ms/btg per ml, p = 0.001). (J Am Coil Cardiol 1995;25:989-94)

Left ventricular function has been reported (1-3) to be an used to study the pharmacodynamic effects of this antiarrhyth- independent predictor of antiarrhythmic drug elficacy in pa- mic drug (17,18). Accordingly, the present study sought to tients with ventricular tachycardia and ventricular fibrillation. 1) evaluate the effects of left ventricular dysfunction on In the setting of left ventricular dysfunction, the effects of quinidine pharmacodynamics; 2) determine whether the antiarrhythmic drugs may potentially be modified by underly- signal-averaged ECG could be used to noninvasively study ing structural heart disease (3-8), electrolyte abnormalities antiarrhythmic drug pharmacodynamics; and 3) determine (9,10), hemodynamic factors (11,12) or neurohumoral mecha- whether differences in quinidine pharmacodynamics predict nisms (13,14) associated with heart failure syndrome. The suppression of inducible ventricular tachycardia. antiarrhythmic drug quinidine has been used frequently for the management of atrial and ventricular tachyarrhythmias in patients with ventricular dysfunction (15,16). Whether quini- Methods dine pharmacodynamics are altered in the presence of left Study protocol. Patients receiving quinidine for treatment ventricular dysfunction is unknown. Because quinidine pro- of supraventricular or ventricular tachyarrhythmias were eligi- longs QRS duration and the QT interval on the surface ble for participation in the study. All patients gave written electrocardiogram (ECG), the signal-averaged ECG could be informed consent to participate in the study, which was approved by the Conjoint Medical Ethics Committee of the University of Calgary. Quinidine therapy was initiated with From the Divisionof Cardiology.,Foothills Medical Centre and Universityof Calgary, Calgary, Alberta, Canada. This study was supported by grants-in-aid quinidine bisulfate, and the dose was titrated to achieve a from the Heart and Stroke Foundation of Alberta and the Medical Research minimal concentration of 10 /xmol/liter or the maximal well Council of Canada (PGl1188), Ottawa, Ontario. Dr. Gillis is a Scholar and Dr. tolerated dose (19). The quinidine bisulfate dose ranged from Duff a Medical Scientist of the Alberta Heritage Foundation for Medical Research, Edmonton, Alberta, Canada. 250 to 750 mg every 8 h. Quinidine was administered at 7 AM Manuscript receivedJune 20, 1994; revisedmanuscript received October 20. on the day of study after quinidine therapy had reached steady 1994, accepted December5, 1994. state (->5 half-lives). Patients then underwent a 24-h drug Address for correspondence: Dr. Anne M. Gillis, Department Medicine, University.of Calgary, 3330 Hospital Drive NW, Calgary.,Alberta, Canada T2N washout period. At 0, 1, 2, 4, 6, 8, 10, 12, 14 and 24 h after the 4N1. last quinidine dose, signal-averaged ECGs were obtained, and

©1995 by the American College of Cardiology 0735-1097/95/$9.5(} 0735-1097(94)00534-W 990 GILL1S ET AL. JACC Vol. 25, No. 5 QUINIDINE PHARMACODYNAMICS April 1995:989-94 plasma was collected for quinidine concentration analysis. Table 1. Clinical Characteristics of 22 Study Patients Rhythm strips were obtained simultaneously for documenta- LVEF <0.35 LVEF ->0.35 tion of the RR interval. (n = 11) (n = 11) Signal-averaged ECG intervals. Bipolar X, Y and Z ECG Age (yr) 64 + 4 62 _+ 12 leads were recorded and analyzed with the Predictor System Men/women 11/0 7/4 (Corazonix Corp.). The ECG signal was digitized at a fre- Cardiac diagnosis quency of 1,000 Hz, and samples were acquired to a noise end Coronary artery disease 9 4 Cardiomyopathy l 2 point of -<0.6 /~V. The unfiltered QRS and QT intervals Valvular heart disease 1 1 recorded at each sampling time were manually measured by Primary electrical 0 4 one of the investigators (M.M.) who had no knowledge of the Presenting arrhythmias identity of the patient and the time of ECG recording. The Ventricular tachycardia 8 8 QRS duration was measured from the onset of the initial high Ventricular fibrillation 3 1 frequency deflection of the QRS complex to the final return of Supraventricular tachycardia 0 2 IVCD 7 l* the signal to baseline. The terminal component of the QT NYHA functional class interval was measured as the point where a tangent drawn from [-ll 10 2J the T wave intersected the isoelectric line (20). The QT III-IV 0 0 interval was corrected for heart rate using the Bazett formula LVEF 26 _+ 7 52 _+ 4t (21). The QRS and QT intervals were remeasured by the same Daily quinidine dose (mg) 1,959 _+ 528 1,639 _+ 636 investigator in 10 patients on separate days, and intraobserver *p < 0.05. ~p < 0.01. Data presented are mean value _+ 1 SD or number of variability was assessed by linear regression analysis. patients. IVCD = intraventricular conduction delay on 12-1ead electrocardio- gram; LVEF - left ventricular ejection fraction; NYHA = New York Heart Quinidine antiarrhythmic elficacy. Patients with inducible Association. sustained ventricular tachycardia during a baseline electro- physiologic study underwent a second electrophysiologic study for assessment of quinidine antiarrhythmic drug etficacy. The Results stimulation protocol has been described elsewhere (8). Anti- Study patients. The clinical characteristics of the 22 study arrhythmic etficacy was defined as the failure to induce >15 patients are shown in Table 1. Patients with a depressed left repetitive ventricular responses with the entire ventricular ventricular ejection fraction more frequently had an intraven- pacing protocol. The electrophysiologic study was performed tricular conduction delay on the 12 lead ECG (p < 0.05) and 5 to 6 h after the last quinidine dose in all but one patient in more frequently had symptoms of congestive heart failure (p < whom the study was performed 2 h after the last quinidine 0.01) than those with a preserved left ventricular ejection dose. The quinidine washout study was performed in these fraction. patients on the day after their electrophysiologic study. Quinidine and metabolite concentrations. Mean quini- Data analysis. Patients were stratified into two groups on dine, 3-hydro~quinidine and dihydroquinidine plasma levels the basis of left ventricular ejection fraction, which was mea- for the low and high left ventricular ejection fraction groups sured by a gated radionuclide angiographic technique: left are shown in Figure 1. Quinidine and metabofite concentra- ventricular ejection fraction <0.35 and ->0.35. Quinidine and tions measured over time and the area under the concentra- metabolite plasma concentrations were determined using a high performance liquid chromatographic assay (22,23). The Figure 1. Quinidine, 3-hydroxyquinidine(3-HO-quinidine) and dihy- trapezoidal rule was used to calculate the area under the droquinidine (DH-quinidine) concentrations over time in the low plasma drug concentration-time curve. Concentrations of (solid symbols, n = 11) and high (open symbols, n = 1l) ejection quinidine, the contaminant dihydroquinidine and the metabo- fraction groups. Data are mean value _+ 1 SD. lite 3-hydroxyquinidine, as well as QRS durations and cor- E rected QT (QTc) intervals, were compared over time between m .•E5.0 the two groups. Quinidine concentration-effect relations were $ T IT e,o QUlNIDINE -,,.o ~Co T I 1 I T ..o 3-.O-OU,N,DINE- 0._ determined by linear regression analyses (24,25). Multiple Z _o A A DH-QUINIDINE rn linear regression analysis was applied to determine the relative I- < o t.r- effects of quinidine on QRS duration and the QTc interval Z (26). The slopes of the concentration-effect relations were IMZ~2.5 -2.0___ o Z -4 compared between the two groups and in patients with induc- o o ible sustained ventricular tachycardia who responded to quin- -4 LU Z Z idine therapy versus those who did not. e", _z Statistical analysis, Data are presented as mean value _+ 1 0 SD. Comparisons between groups were made using the non- o 0 6 12 18 24 3 paired t test or Fisher exact test, as appropriate. TIME (hrl JACC Vol. 25, No. 5 G1LLIS ET AL. 991 Aprll 1995:989-94 QUINIDINE PHARMACODYNAMICS

Table 2. Quinidine Pharmacokinetics and Pharmacodynamics and did not change significantly over time (data not shown). LVEF <0.35 LVEF >-0.35 The maximal change (from peak to trough effect) in QTc (n = 11) (n = 11) intervals observed during the washout phase correlated lin- AUC quinidine (tzg/ml'h) 64.7 + 18.1 55.4 +_ 17.8 early, whereas that for QRS durations correlated inversely, AUC 3-HQ 0xg/ml.h) 16.0 _+ 6.8 12.1 + 4.2 with left ventricular ejection fraction (p < 0.05) (Fig. 2). AUC DHQ (~g/ml.h) 3.1 + 3,2 2.7 +_ 3.3 Linear concentration-effect relations. Linear quinidine K~,I quinidine (h 1) 0.0511 - 0,0113 0.{)554 + 0.0116 concentration-effect relations for QRS durations were signifi- tV2 quinidine (h) 14.3 +_ 3,4 13.0 + 2.8 cant (p < 0.05) in all ll patients with a low left ventricular Slope QRS (ms//xgper ml) 451 + 2.32 4.64 _+ 2.74 ejection fraction and in 9 of 11 with a high left ventricular (2A0-10.9) (1.07-11.8) ejection fraction. For QTc intervals, linear concentration-effect r value 0.83 "- 0.08 0.74 _+ 0.11 (0.68-0.94) (0.59-0.91) relations were significant (p < 0.05) in 9 of i1 patients with a Slope QTc (ms/gg per ml) 15.7 + 9.7 29.5 -+ 11.2" low left ventricular ejection fraction and 10 of 11 with a high (3.81-22.8) (9.9-43.2) left ventricular ejection fraction (Fig. 3). The slopes of the r value 0,68 -+ 0.22 0.76 + 0.08 QRS concentration-effect relations were similar in both groups (0.38-0.90) (0.62-0.85) (Table 2); however, slopes describing changes in QTc intervals *p - 0.001. Data presented are mean value _+ I SD (range). AUC = area were significantly lower in the low than in the high left under concentration-time curve; DHQ = dihydroquinidine; 3-HQ = 3-hy- ventricular ejection fraction group (p = 0.001) (Table 2). droxyquinidine; K~l eliminationrate constant; LVEF = left ventricular Similar results were observed when 3-hydroxyquinidine con- ejection fraction; QTc = corrected QT interval; Slope - slope of linear centration-effect relations were analyzed (data not shown). In regression analysis; t]A, - eliminationhalf-life. addition, a significant linear correlation was observed between the slopes of the quinidine concentration-effect relations de- tion-time curves were similar in both groups (Table 2), as was scribing changes in the QTc interval and left ventricular the elimination half-life of quinidine. ejection fraction (p < 0.001) (Fig. 3). Multiple linear regres- Quinidine pharmacodynamics. Intraobserver variability of sion analysis revealed that quinidine concentration correlated QRS and QT measurements was minimal. The correlation independently with QRS duration in six patients in the low coefficient for repeat QRS measurements on a different day ejection fraction group and in two in the high ejection group, was 0.87, the slope 1.03 and intercept -5 ms. The correlation whereas quinidine concentration correlated independently coefficient for repeat QT measurements was 0.95, the slope with the QTc interval in one patient in the low ejection fraction 1.08 and intercept 9 ms. Mean QRS durations and QTc group and in four in the high ejection fraction group. In the intervals over the 24-h washout period are shown in Figure 2. remaining patients, QRS durations and QTc intervals were not The QRS durations and the QTc intervals were greater in the independently associated with quinidine concentration low than in the high left ventricular ejection fraction group (p = NS). throughout the washout period (p < 0.05). The RR intervals Antiarrhythmic drug efficacy. During quinidine therapy, were similar during the 24-h quinidine washout in both groups ventricular tachycardia could not be induced during a follow-up electrophysiologic study in 2 of 10 patients in the low left ventricular ejection fraction group and in 6 of 9 in the high left Figure 2. Left, QRS durations and corrected QT intervals (QTc) ventricular ejection fraction group (p < 0.07). Quinidine during 24-h quinidine washout period for the low (solid circles, n = concentrations measured at the time of the electrophysiologic 11) and high (open circles, n = 11) left ventricular ejection fraction study were 2.7 + 1.3/zg/ml in the quinidine drug responders (LVEF) groups. Data are mean value _+ l SD. Right, Relations and 3.5 + 0,7 p,g/ml in the nonresponders (p = NS). The slopes between left ventricular ejection fraction and maximal change (A) in QRS duration and corrected QT interval (QTc) measured during the of the quinidine concentration-effect relations describing study. changes in the QTc interval were significantly higher in the quinidine drug responders (24.9 _+ 10.0 ms/p,g per ml) than in r = -0.50 180 ~ 40 p < 0.05 nonresponders (15.9 + 7.1 ms/#g per ml, p < 0.05). Differ- ences in the slopes of the quinidine concentration-effect rela- E ,,~ 125 tions describing changes in ventricular conduction time were e~ not observed between drug responders (4.73 _+ 2.17 ms/t~g per 0 7O ~ 0 ml) and nonresponders (5.03 _+ 3.52 ms//zg per ml, p = NS).

.~150 r = 0.40 • 525 p < 0.05 • • ~ 100 • • Discussion o 450 Although it is well recognized that left ventricular dysfunc- 0 tion is an independent predictor of antiarrhythmic drug inef- 375 ~ 0 t , t ," i , i , i ~ i i ficacy (1-3), few studies have evaluated the influence of left 0 6 12 18 24 0 0.40 0.80 ventricular dysfunction on the pharmacodynamics of anti- TIME (hr) LVEF arrhythmic drugs. The major findings of the present study are 992 GILLIS ET AL. JACC Vol. 25, No. 5 QUINID1NE PHARMACODYNAMICS April 1995:989-94

r = -0.31 12 p=N$ • '~ 120 | Figure 3. Examples of linear concentration-effectrela- tions. Top left, Linear relations between the QRS interval ®-36 i % li 100 and quinidine concentration for one representative pa- re" .a=" 12I ~n~ di I tient in the high (open circles [slope 3.28 ms/~g per ml, 80 r - 0.89, p < 0.001D and one in the low (solid circles [slope 3.01 ms//zg per ml, r = 0.90, p < 0.001]) left ventricular ejection fraction group. Bottom left, Linear ""d 500 O ~ = p < 0.001 O relations between the corrected QT interval (QTc) and 50 I r = 0.710Og~..B i O quinidine concentration for both patients (high left ven- E • . tricular ejection fraction, slope 42.8 ms//xg per ml, r = u 400 I-- °* 0.84, p = 0.003; low left ventricularejection fraction,slope @ 18.9 ms/~g per ml, r = 0.87, p = 0.001). Right, Relations 3O0 o between slopes of the concentration-effectrelations and i I i t i left ventricular ejection fraction. 3.5 7.0 0 0.40 0.80 QUINIDINE {~g/rnl} LVEF that quinidinc concentration-effect relations are altered in the effects of the drugs on left ventricular function or by the effects setting of left ventricular dysfunction and that these altered of left ventricular dysfunction on cardiac repolarization. 2) concentration-effect relations are associated with antiarrhyth- Increased sympathetic activity or increased circulating cat- mic drug inefficacy. The effect of quinidine on ventricular echolamines associated with clinical congestive heart failure repolarization and conduction time correlated directly and might attenuate or reverse the effects of antiarrhythmic drugs, inversely, respectively, with left ventricular ejection fraction. including quinidine (30-35). Against this hypothesis, none of Quinidine pharmacodynamics. Pharmacodynamic models our patients with a low left ventricular ejection fraction had are used to describe the equilibrium relation between concen- functional class III or IV congestive heart failure symptoms. tration and effect (24). In the present study, linear quinidine 3) Ion channel function may be altered in the setting of concentration-effect relations were observed for changes in myocardial disease (4,7,36,37). It is possible that outward QRS durations and QT intervals. This is not surprising because potassium currents in hypertrophied/failing myocytes may be the concentration range evaluated in the present study was less responsive to the effects of quinidine. 4) Changes in small and well below the concentration at which 50% of plasma protein binding associated with significant left ventric- maximal effect would be observed, and therefore it fell on the ular function might alter the unbound fraction of quinidine linear portion of the concentration-response relation for quin- and the pharmacodynamics of quinidine (38,39). idine (24,25). The slope of the linear regression analysis We also demonstrated a negative linear correlation be- describes the relation between concentration and effect and tween maximal change in QRS duration and left ventricular allows comparison of these relations when concentrations vary ejection fraction. These findings might be explained by the between subjects (24,25). A significant difference in the quin- presence of an intraventricular conduction abnormality in idine concentration-effect relations describing changes in the seven of the patients in the low left ventricular ejection fraction QTc interval was observed between patients with a low and group. Diseased, slowly conducting myocardium might be those with a high left ventricular ejection fraction. Because the more vulnerable to conduction slowing or blocking by quini- slope of the concentration-effect relation is a derived value, dine and other antiarrhythmic drugs because of altered cellular errors in this calculation might explain the stronger correlation electrophysiology or abnormalities of cell-to-cell coupling observed between slope and left ventricular ejection fraction (4,5,7,40,41). than that observed between the maximal change in the QTc Quinidine antiarrhythmic efficacy. In patients with ventric- interval and left ventricular ejection fraction. However, this ular tachycardia we observed that quinidine concentration- calculation incorporates differences in plasma quinidine con- effect relations describing changes in ventricular repolarization centrations observed between patients and hence could be a correlated with antiarrhythmic efficacy. We (42) and other stronger measure of the effects of quinidine than changes in investigators (43) previously observed that greater antiarrhyth- ECG intervals alone. mic drug-induced prolongation of ventricular refractoriness is We previously reported (11) similar relations between the associated with antiarrhythmic efficacy. Patient responders had pharmacodynamics of disopyramide and left ventricular ejec- a higher left ventricular ejection fraction, and many did not tion fraction. There are four potential explanations for these have coronary artery disease. However, future studies will be relations: 1) Diastolic stretch is well known to modulate action necessary to determine the mechanism(s) of the altered phar- potential duration and ventricular refractory periods (27-29). macodynamics in the setting of left ventricular dysfunction. Thus, direct electrophysiologic effects of antiarrhythmic drugs Potential limitations. The ECG intervals measured in the may be partially offset by the indirect consequences of the present study reflect global ventricular depolarization and JACC Vol. 25, No. 5 GILLtS ET AL. 993 April 1995:989 94 QUINIDINE PHARMACODYNAMICS repolarization. Thus, the effects of quinidine on global electro- Prognostic importance of the serum magnesium concentration in patients with congestive heart failure. J Am Coil Cardiol 1990;16:827-31. physiologic variables may not reflect the effects of quinidine on 11. Duff HJ, Mitchell LB, Nath CF, Manyari DE, Baynton R, Wyse DG. local tissue involved in a reentrant circuit. As well, the non- Concentration-response relationships of disopyramide in patients with ven- steady state conditions that existed during the present study tricular tachycardia. Clin Pharmacol Ther 1989;45:542-7. may not reflect the concentration-effect relations that exist 12. Hansen DC, Borganelli M, Stacy GP Jr, Taylor LK. Dose-dependent inhibition of stretch-induced arrhythmias by gadolinium in isolated canine during the steady state. Measurement of the QT interval may ventricles. Evidence for a unique mode of antiarrhythmic action. Circ Res be confounded by the presence of a U wave. In the present 1991;69:820-31. study we attempted to minimize this problem by selecting a 13. Bashir Y, Sneddon JF, O'Nunain S, et al. Comparative electrophysiological lead with one discrete waveform (20). Furthermore, all mea- effects of captopril or hydralazine combined with nitrate in patients with left ventricular dysfunction and inducible ventricular tachycardia. Br Heart J surements were performed by one observer who was unaware 1992;67:355-60. of patient identity and time of ECG recording. The QT interval 14. Poquet F, Ferguson J, Rouleau JL. The antiarrhythmic effect of the ACE is most frequently corrected for heart rate using the Bazett inhibitor catopril in patients with congestive heart failure largely is due to its formula (21). Nevertheless, there is controversy concerning potassium sparing effects. Can J Cardiol 1992;8:589-95. 15. Mahmarian J J, Verani MS, Hohmann T, et al. The hemodynamic effects of which formula is appropriate to correct for heart rate (44). The sotalol and quinidine: analysis by use of rest and exercise gated radionuclide use of Bazett formula may cause a substantial overcorrection, angiography. Circulation 1987;76:324-31. particularly at slow heart rates. However, in the present study 16. Sheldon RS, Duff HJ, Mitchell LB, Wyse DG, Manyari DE. Effect of oral heart rates were similar throughout the 24-h washout period in combination therapy with mexilitine-quinidine therapy on left and right ventricular function. Am Heart J 1988;115:1030-6. both patient groups, thus minimizing this limitation. Moreover, 17. Wechsler ME, Steinberg JS, Giardina E-GV. Time course of moricizine's the use of the Fridericia formula (44) did not change the effect on signal-averaged and 12 lead electrocardiograms: insight into results (unpublished data). mechanism of action. J Am Coil Cardiol 1991;17:1626-33. Summary. We observed altered quinidine pharmacody- 18. Freedman RA, Steinberg JS. Selective prolongation of QRS late potentials by sodium channel blocking antiarrhythmic drugs: relation to slowing of namics in patients with significant left ventricular dysfunction. ventricular tachycardia. J Am Coll Cardiol 1991;17:1017-25. The effects of quinidine on ventricular repolarization directly 19. Duff HJ, Wyse DG, Manyari D, Mitchell LB. Intravenous quinidine: correlate with left ventricular ejection fraction. Furthermore, relations among concentration, tachyarrhythmia suppression and electro- these effects are associated with antiarrhythmic efficacy for physiologic actions with inducible sustained ventricular taehycardia. Am J Cardiol 1985;55:92-7. ventricular tachycardia. However, the mechanism(s) of this 20. Garson A. How to measure the QT interval--what is normal? Am J Cardiol altered pharmacodynamic response requires further study. 1993;72:14B-6B. 21. Bazett HC. An analysis of the time-relationship of electrocardiograms. Heart 1920;7:353-70. We thank Linda Ellis, BSc and Heather J. Mathison, BSc for technical assistance 22. Drayer DE, Restivo K, Reidenberg MM. Specific determination of quinidine and Marilyn Devlin for manuscript preparation. and (3S)-3-hydroxyquinidine in human serum by high pressure liquid chromatography. J Lab Clin Med 1977;90:816-22. 23. Kavanagh KM, Wyse DG, Mitchell LB, Gilhooly T, Gillis AM, Duff HJ. Contribution of quinidine metabolites to electrophysiologic responses in References human subjects. Clin Pharmacol Ther 1989;46:352-8. 24. Holford NHG, Sheiner LB. Understanding the dose-effect relationship: 1. Spielman SR, Schwartz JS, McCarthy DM, et al. Predictors of the success or clinical application of pharmacokinetic-pharmacodynamic models. Clin failure of medical therapy in patients with chronic recurrent sustained Pharmacol 1981;6:429-53. ventricular tachycardia: a discriminant analysis. J Am Coll Cardiol 1983;1: 25. Gillis AM, Duff HJ, Mitchell LB, Wyse DG. Myocardial uptake and 401-8. pharmacodynamics of procainamide in patients with coronary heart disease 2. Hohnloser SH, Raeder EA, Podrid PJ, Graboys TB, Lown B. Predictors of and sustained ventricular tachyarrhythmias. J Pharmacol Exp Ther 1993;266: antiarrhythmic efficacy in patients with malignant ventricular tachyarrhyth- 1001-6. rajas. Am Heart J 1987;114:1-7. 26. Zar JH. Multiple regression and correlation. In: McElroy WD, Swanson CP, 3. Poole JE, Mathisen TL, Kudenchuk PJ, ct al. Long-term outcome in patients editors. Biostatistical Analysis. Englewood Cliffs (NJ): Prentice-Hall, 1974: who survive out of hospital ventricular fibrillation and undergo electrophysi- 252- 80. ologic studies: evaluation by electrophysiologic subgroups. J Am Coil 27. Reiter MJ, Synhorst DP, Mann DE. Electrophysiological effects of acute Cardiol 1990;16:657-65. ventricolar dilatation in the isolated rabbit heart. Circ Res 1988;62:554-62. 4. Zeiler RH, Gough WB, EI-Sherif N. Electrophysiologic effects of 28. Calkins H, Maughan WL, Weisman HF, Suguira S, Sagawa K, Levine JH. propafenone on canine ischemic cardiac cells. Am J Cardiol 1984;54:424-9. Effect of acute volume load on refractoriness and arrhythmia development in 5. Gillis AM, Lester WM, Keashly R. Propafenone disposition and pharmaco- isolated, chronically infarcted canine hearts. Circulation 1989;79:687-97. dynamics in normal and norepinephrine-induced cardiomyopathic rabbit 29. Dean JW, Lab MJ. Regional changes in ventricular excitability during load hearts. J Pharmacol Exp Ther 1991;258:722-7. manipulation of the in situ pig heart. J Physiol 1990;429:387-400. 6. Myerburg RJ, Kessler KM, Castellanos A. Sudden cardiac death: structure, 30. Saul JP, Arai Y, Berger RD, Lilly LS, Colucci WS, Cohen RJ. Assessment of function and time-dependence of risk. Circulation 1992;85 Suppl I:I-2-10. autonomic regulation in chronic congestive heart failure by heart rate 7. Montero M, Beyer T, Schmitt C, Kubler W, Brachmann J. Differential effects spectral analysis. Am J Cardiol 1988;61:1292-9. of quinidine on transmembrane action potentials of normal and infarcted 31. Ellenbogen KA, Mohanty PK, Szentpetery S, Thames MD. Arterial barore- canine Purkinje fibers. J Cardiovasc Pharmacol 1992;20:304-10. flex abnormalities in heart failure: reversal after orthotopic cardiac trans- 8. Hii JTY, Traboulsi M, Mitchell LB, Wyse DG, Duff H J, Gillis AM. Infarct plantation. Circulation 1989;79:51-8. artery patency predicts outcome of serial electropharmacological studies in 32. Dibner-Dunlap ME, Thames MD. Control of sympathetic nerve activity by patients with malignant ventricular tachyarrhythmias. Circulation 1993;87: vagal mechanoreflexes is blunted in heart failure. Circulation 1992;86:1929- 764 -72. 34. 9. Podrid PJ. Potassium and ventricular arrhythmias. Am J Cardiol 1990;65: 33. Kienzle MG, Ferguson DW, Birkett CL, Myers GA, Berg WJ, Mariano DJ. 33E-44E. Clinical, hemodynamic and sympathetic neural correlates of heart rate 10. Gottlieb SS, Baruch L, Kukin ML, Bernstein JL, Fisher ME, Packer M. variability in congestive heart failure. Am J Cardiol 1992;69:761-7. 994 GILLIS ET AL. JACC Vol. 25, No. 5 QUIN1DINE PHARMACODYNAMICS April 1995:989-94

34. Morady F, Kou WH, Kadish AH, et al. Antagonism of quinidine's electro- 40. De Langen CDJ, Hanich RF, Michelson EL, et al. Differential effects of physiologic effects by epinephrine in patients with ventricular tachycardia. procainamide, lidocaine and acetylstrophanthidin on body surface potentials J Am Coil Cardiol 1988;12:388-94. and epicardial conduction in dogs with chronic myocardial infarction. J Am 35. Markel ML, Miles WM, Luck JC, Klein LS, Prystowsky EN. Differential effects Coll Cardiol 1988;11:403-13. of isoproterenol on sustained ventricular tachycardia before and during procain- 41. Kieval RS, Spear JF, Moore EN. Gap junction conductance in ventricular amide and quinidine antiarrhythmic therapy. Circulation 1993;87:783-92. myocyte pairs isolated from postischemic rabbit myocardium. Circ Res 36. Kleiman RB, Houser SR. Outward currents in normal and hypertrophied 1992;71:127-36. feline ventricular myocytes. Am J Physiol I989;256:H1450-61. 42. Gillis AM, Wyse DG, Duff H J, Mitchell LB. Drug response at electrophar- 37. Xu X, Best PM. Decreased transient outward K ' current in ventricular macologic study in patients with ventricular tachyarrhythmias: the impor- myocytes from acromegalic rats. Am J Physiol 1991;260:H935-42. tance of ventricular refractoriness. J Am Coll Cardiol 1991;17:914-20. 38. Kessler KM, Kissane B, Cassidy J, Pefkaros K, Humburg C, Myerburg RJ. 43. Furukawa T, Rozanski J J, Moroe K, Gosselin AJ, Lister JW. Efficacy of Dynamic variability of binding of antiarrhythmic drugs during the evolution procainamide on ventricular tachycardia: relation to prolongation of refrac- of acute myocardial infarction. Circulation 1984;70:472-8. toriness and slowing of conduction. Am Heart J 1989;118:702-8. 39. Gillis AM, Kates RE. Influence of protein binding on the myocardial uptake and 44. Funck-Brentano C, Jaillon P. Rate-corrected QT interval: techniques and pharmacodynamics of propafenone. J Cardiovasc Pharmacol 1986;8:1163-7. limitations. Am J Cardiol 1993;72:17B-22B.